Figure 2 | Scientific Reports

Figure 2

From: Using human in vitro transcriptome analysis to build trustworthy machine learning models for prediction of animal drug toxicity

Figure 2

Results from BUN level prediction ML comparative regression analysis. Bar charts showing the RMSE and MAE scores for the test datasets (using best parameters) and the mean MAE training scores after 10-fold cross validation with the standard deviation shown as error bars. Left y-axis is used for MAE/RMSE scores. Line plots show the R2 score for test datasets (using best parameters) with the right y-axis used for R2 scores. For Linear Regression, SVM, KNN, XGBoost and GP tSVDh dimensionality reduction was used whereas for other approaches it was not since resulting models were more predictive without it.

Back to article page